observations of radiocarbon in co2 at la jolla, california...

Observations of radiocarbon in CO2 at La Jolla, California, USA1992–2007: Analysis of the long-term trend

Heather D. Graven,1,2 Thomas P. Guilderson,3,4 and Ralph F. Keeling1

Received 11 July 2011; revised 9 November 2011; accepted 16 November 2011; published 25 January 2012.

[1] High precision measurements of D14C were performed on CO2 sampled at La Jolla,California, USA over 1992–2007. A decreasing trend in D14C was observed, whichaveraged �5.5 ‰ yr�1 yet showed significant interannual variability. Contributions to thetrend in global tropospheric D14C by exchanges with the ocean, terrestrial biosphereand stratosphere, by natural and anthropogenic 14C production and by 14C-free fossilfuel CO2 emissions were estimated using simple models. Dilution by fossil fuelemissions made the strongest contribution to the D14C trend while oceanic 14C uptakeshowed the most significant change between 1992 and 2007, weakening by 70%.Relatively steady positive influences from the stratosphere, terrestrial biosphere and 14Cproduction moderated the decreasing trend. The most prominent excursion from the averagetrend occurred when D14C decreased rapidly in 2000. The rapid decline in D14C wasconcurrent with a rapid decline in atmospheric O2, suggesting a possible cause may be theanomalous ventilation of deep 14C-poor water in the North Pacific Ocean. We additionallyfind the presence of a 28-month period of oscillation in the D14C record at La Jolla.

Citation: Graven, H. D., T. P. Guilderson, and R. F. Keeling (2012), Observations of radiocarbon in CO2 at La Jolla, California,USA 1992–2007: Analysis of the long-term trend, J. Geophys. Res., 117, D02302, doi:10.1029/2011JD016533.

1. Introduction

[2] Long term atmospheric measurements of radiocarbon,14C, in CO2 began in 1954 at Wellington, New Zealand[Rafter, 1955]. Since then, observations from New Zealand[Rafter and Fergusson, 1957; Manning et al.,1990; Currieet al., 2009], Norway [Nydal and Lövseth, 1965, 1983],central Europe [Levin et al., 1985; Levin and Kromer, 2004]and other sites have recorded large changes in the 14C/Cratio of CO2. In the 1950s and 1960s, testing of nuclearweapons added an excess of 14C atoms that approximatelydoubled the atmospheric inventory of 14C. As the testingceased and bomb-derived 14C entered the oceanic and ter-restrial carbon reservoirs, observations of 14C/C in CO2

revealed a quasi-exponential decline that enabled investiga-tion of mixing rates between different parts of the atmo-sphere and exchange rates between the atmosphere and theocean and terrestrial ecosystems [e.g., Rafter and Fergusson,1957; Lal and Rama, 1966; Goudriaan, 1992; Hesshaimeret al., 1994; Trumbore, 2000; Naegler et al., 2006].

[3] Presently, 14C exchanges between the atmosphere andthe ocean and terrestrial biosphere are redistributing thebomb-derived excess 14C from short term to longer termreservoirs. 14C exchanges are also responding to the dilutionof atmospheric 14C by fossil fuel-derived CO2 which has no14C because of radioactive decay [Suess, 1955; Keeling,1979; Tans et al., 1979; Stuiver and Quay, 1981]. Since theresponse of land and ocean carbon reservoirs to these per-turbations in 14C/C is governed by the same exchange pro-cesses that determine anthropogenic CO2 uptake and storage,continued observation and understanding of 14C dynamicscan provide constraints on terrestrial and oceanic carboncycling and the potential magnitude and sustainability ofCO2 sinks [Levin and Hesshaimer, 2000; Randerson et al.,2002].[4] 14C/C ratios can also be used to identify local additions

of CO2 from fossil fuel emissions by observation of 14Cdilution in comparison to background air [e.g., Tans et al.,1979; Levin et al., 1989; Meijer et al., 1996; Turnbullet al., 2006; Levin and Rödenbeck, 2008]. Quantification offossil fuel-derived CO2 can be useful for resolving budgets ofCO2 contributions from industrial versus biospheric or oce-anic sources [e.g., Turnbull et al., 2006;Graven et al., 2009],for detecting temporal or spatial patterns in fluxes of differenttypes [e.g., Hsueh et al., 2007; Turnbull et al., 2009a], or forestimating fossil fuel emissions within a catchment area [e.g.,Levin et al., 2003; van der Laan et al., 2010; Turnbull et al.,2011]. The use of atmospheric observations to estimate fossilfuel emissions promises to become important as emissions ofCO2 and other greenhouse gases are more heavily regulatedand require independent verification of economic data-based

1Scripps Institution of Oceanography, University of California, SanDiego, La Jolla, California, USA.

2Institute of Biogeochemistry and Pollutant Dynamics, ETH Zurich,Zurich, Switzerland.

3Center for Accelerator Mass Spectrometry, Lawrence LivermoreNational Laboratory, Livermore, California, USA.

4Department of Ocean Sciences, University of California, Santa Cruz,California, USA.

Copyright 2012 by the American Geophysical Union.0148-0227/12/2011JD016533

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 117, D02302, doi:10.1029/2011JD016533, 2012

D02302 1 of 14

http://dx.doi.org/10.1029/2011JD016533

inventories [Nisbet, 2005; Marquis and Tans, 2008]. Mea-surements that resolve variability in background air areessential to this technique since uncertainty in background aircomposition limits the precision of observation-based esti-mates of fossil fuel-derived CO2 [Graven et al., 2009;Turnbull et al., 2009b].[5] Measurements of 14C/C ratios are typically referenced

to the Modern Standard and reported as D14C, whichincludes a correction for radioactive decay between sam-pling and analysis and a correction for mass-dependentfractionation using measurements of 13C/12C in the sample[Stuiver and Polach, 1977]. Use ofD14C notation eliminatesthe effect of fractionating processes such as photosyntheticassimilation or oceanic CO2 uptake on the 14C/C ratio inCO2. D14C in CO2 is therefore sensitive to natural andanthropogenic 14C production as well as to exchanges ofcarbon with reservoirs that have a different D14C signature.[6] Here we present D14C measurements in CO2 samples

from La Jolla, California, USA collected at roughly monthlyintervals between 1992 and 2007. In this paper, we focus onanalyzing the D14C trend over the 16-year record. Wecompare the observed trend to simple models of the con-tributions to the global tropospheric trend in D14C between1992 and 2007 from the release of fossil fuel CO2, naturaland anthropogenic production of 14C and carbon exchangeswith the stratosphere, ocean, and biosphere. We then eval-uate interannual variability in the trend of D14C. In theaccompanying paper, we present D14C observations from 6other global sites for 2- to 9-year periods ending in 2007 andexamine spatial gradients and seasonal cycles [Graven et al.,2012].

2. Methods

[7] Atmospheric flask sampling at La Jolla, California isconducted at the Scripps Pier (32.87°N, 117.25°W) whenmeteorological conditions are favorable for collecting clean,marine air and avoiding local contamination. Such condi-tions are met when strong, stable winds originating from thesouthwesterly sector (offshore) are present and a low, stableCO2 concentration is identified with a continuous CO2

analyzer.[8] The representativeness of air collected at La Jolla

under clean air conditions can be assessed by comparingobserved CO2 concentrations with other background sites.Observed annual mean CO2 gradients are less than �0.2ppm between La Jolla and the Pacific Ocean Station at 30°N,122.85°W [Conway and Tans, 2004] and less than �0.5ppm between La Jolla and other Scripps CO2 stations inclosest proximity (Kumukahi, Hawaii and Point Barrow,Alaska) [Keeling and Piper, 2001; Keeling et al., 2011]. TheCO2 concentration in the air collected at La Jolla is slightlyhigher than Kumukahi and slightly lower than Point Barrow,in accordance with the observed meridional CO2 gradient inbackground air [Keeling and Piper, 2001; Masarie andTans, 1995; Keeling et al., 2011]. The seasonal cycle ofD14C at La Jolla, presented in the accompanying paper, alsosupports the representativeness of clean, marine air sampledunder these conditions. If local contamination stronglycontributed to the seasonal cycle of D14C at La Jolla, theminimum in D14C would be expected to occur in the fall

months when polluted continental air is transported offshoremost frequently [Conil and Hall, 2006; Riley et al., 2008]. Infact, the maximum D14C occurs in the fall months at LaJolla, consistent with other Northern Hemisphere observa-tions of background air [Graven et al., 2012; Levin andKromer, 2004; Turnbull et al., 2007].[9] Evacuated 5 L spherical glass flasks are sampled by

opening a single ground taper joint stopcock sealed withApiezon® grease and filling with whole air. At La Jolla, sixflasks are sampled concurrently, while at other Scripps CO2

stations, 2–3 flasks are sampled concurrently. Flask air isdried and the CO2 concentration is measured by non-dis-persive infrared gas analysis [Keeling et al., 2002] at theScripps laboratory. CO2 is extracted from the remainingflask air by passing 2–4 L of air through a spiral quartz trapimmersed in liquid nitrogen, then the CO2 sample is sealedand stored in a Pyrex® tube. The same flask handling,storage and extraction procedures are used for analysis ofd13C in CO2 at Scripps with measurement precision of <0.03‰ [Guenther et al., 2001]. An archive of such CO2 samplesdating back to July 1992 from La Jolla was available foranalysis. Samples collected between July 1992 and Decem-ber 2007 were analyzed for 14C in CO2, including 79 sampledates for which two or more replicate samples were mea-sured. These CO2 samples were analyzed together with CO2

samples from six other sites [Graven et al., 2012].[10] All CO2 samples were converted to graphite and

analyzed by accelerator mass spectrometry (AMS) at theCenter for Accelerator Mass Spectrometry at LawrenceLivermore National Laboratory (LLNL) between 2003 and2009 [Graven et al., 2007; Graven, 2008]. We utilize theD14C notation implicitly as a geochemical sample withknown age and d13C correction (equivalent to D in work byStuiver and Polach [1977]). Ratios of 14C/C are correctedfor decay between sampling and analysis dates and for massdependent fractionation using d13C. The d13C correctionuses d13C measured in concurrently sampled CO2 [Guentheret al., 2001] to normalize atmospheric samples to the�25‰reference. Unlike some smaller AMS systems which causesignificant fractionation during ionization, there is no evi-dence for fractionation within the LLNL AMS ion source[Proctor et al., 1990]. This is shown by nearly constant14C/13C ratios measured throughout the analysis of individ-ual samples [Fallon et al., 2007]. Slight drifts in 14C/13Cratios can be attributed to stripping efficiency and are suc-cessfully canceled by normalization with reference materi-als. Therefore, in-line d13C correction is not required orperformed at LLNL.[11] Individual measurement uncertainty in D14C is �1.7

‰ for most samples, where uncertainty is determined by thereproducibility of D14C in CO2 extracted from whole airreference cylinders [Graven et al., 2007; Graven, 2008].Measurements conducted prior to 2006 have uncertaintiesbetween �1.7 and �3.3 ‰ [Graven, 2008], since samplehandling and data processing were not yet optimized forCO2 samples. However, drawing on a large archive ofexisting samples allowed the samples to be selected ran-domly for analysis between 2003 and 2009. Samples fromdifferent years and different sites were included in eachmeasurement batch [Graven, 2008], so that differences inuncertainty between measurement batches can be expected

GRAVEN ET AL.: 14CO2 AT LA JOLLA D02302D02302

2 of 14

to have little impact on the features contained in the atmo-spheric records.

3. D14C Observations

[12] Measurements of D14C are shown in Figure 1 andlisted in Appendix A. These data are also available at theScripps CO2 Program Web site: http://scrippsco2.ucsd.edu/.[13] Figure 2a shows seasonally adjusted observations of

D14C at La Jolla to emphasize variations in the data thatoccur at timescales longer than one year. Seasonal cycles arepresented and discussed in the accompanying paper [Gravenet al., 2012]. The seasonal cycles were removed by firstdetrending the data with a cubic smoothing spline with cut-off period of 24 months [Enting, 1987]. Second, a loosecubic smoothing spline was fit to the detrended data (cutoffperiod of 4 months) and finally, the loose spline was sub-tracted from the original observations.[14] D14C at La Jolla decreased by nearly 100 ‰ between

1992 and 2007. A linear least squares fit to all measurementsresults in a slope of �5.5 � 0.1‰ yr�1, where 0.1 is the 1-suncertainty [Cantrell, 2008]. The trend was slightly weakerduring the second half of the observation period; a linear fitto observations between mid-2001 and the end of 2007yields a slope of �5.0 � 0.2 ‰ yr�1 [Graven et al., 2012]while a fit to observations between mid-1992 to mid-2001yields �5.7 � 0.1 ‰ yr�1. We have also fit an exponentialtrend to the observedD14C; the derivative of the exponentialand linear fits are shown in Figure 2b. Compared to theconstant trend of �5.5 ‰ yr�1 computed by the linear fit,the exponential derivative slows from �8‰ yr�1 in 1992 to�3 ‰ yr�1 in 2007.[15] Figure 2b also shows the derivative of the seasonally

adjusted D14C observations. The growth rate of D14C at LaJolla showed large variability over 1992–2007. The stron-gest excursion from the linear or exponential trend was anespecially rapid decrease in D14C in 2000. Local minima inthe trend of D14C are also observed at roughly 2 yearintervals, suggesting a short term periodicity exists in D14C

at La Jolla. Features of variability in the trend are not likelyto be artifacts of the spline fitting technique, since trendscalculated using annual means (Section 5.2) and low-passfiltering (Section 5.3) show similar features.

4. Global Trend in Tropospheric D14C

4.1. Description of Box Model Formulation

[16] In this section, we will describe the box model for-mulation used to simulate influences on the trend in D14C

Figure 1. D14C measured in CO2 sampled at La Jolla with a cubic smoothing spline. Replicate measure-ments have been averaged. Error bars show measurement uncertainty or the standard deviation inD14C ofreplicate samples, whichever is larger.

Figure 2. (a) Seasonally adjusted D14C at La Jolla with acubic smoothing spline. (b) The derivative of the splinecurve from Figure 2a is shown as the solid line. Also shownin Figure 2b are lines representing the derivative of a linear fit(dashed, �5.5 ‰ yr�1) and an exponential fit (dash-dotted).


3 of 14

from the emission of fossil fuel-derived 14C-free CO2, nat-ural and anthropogenic 14C production, and 14C and carbonexchanges between the troposphere and the ocean, the landbiosphere, and the stratosphere. In Section 4.2, we willpresent the results of the model and in Section 5.1, we willdiscuss the results, the limitations of the simple models usedhere, and a comparison with similar calculations performedby Levin et al. [2010].[17] The influence on tropospheric D14C from exchange

with a carbon reservoir is primarily determined by the D14Cdisequilibrium. In order to formulate the most precise esti-mate of these influences, we used the observed troposphericD14C, d13C and CO2 to calculate the evolution of D14C andcarbon and isotopic fluxes in separate, uncoupled forwardmodels of the ocean, biosphere and stratosphere. We thensynthesize the separate model contributions and evaluate thecorrespondence between the simulated and observed D14Ctrend. While individual components are forced by observedatmospheric changes, they are not constrained to add upto the observed overall D14C change in the atmosphere. Acomparison of the sum of the components with observationsthus provides an important consistency check. Estimatesof global averages were formulated by weighted averagesof clean-air station data including results from Neftel et al.[1994] and Keeling and Whorf [2005] for atmospheric CO2

concentrations, from Friedli et al. [1986] and Keeling et al.[2005] for d13C, and from this work and that of Stuiver et al.[1998], Levin and Kromer [2004], Levin et al. [2007] andGraven et al. [2012] for D14C.[18] The box model setup is depicted in Figure 3 and

summarized in Table 1. The biospheric and oceanic

components were initialized with simulations of constantpreindustrial atmospheric composition lasting 30,000 yearsto achieve steady state. Then, simulations of the biosphericand oceanic components were conducted using records ofatmospheric composition beginning in 1511 for D14C[Stuiver et al., 1998] and 1720 and 1744 for CO2 con-centrations and d13C [Neftel et al., 1994; Friedli et al.,1986], respectively. The stratospheric component was sim-ulated beginning in 1900. We present model results for theperiod of observation at La Jolla: 1992 through the end of2007.[19] The decrease in D14C caused by fossil fuel emissions

was estimated by mixing the global annual CO2 emissionsfrom inventories of economic data [Marland et al., 2008;Canadell et al., 2007] into the entire troposphere (78% ofthe atmosphere). We included a 10% uncertainty in reportedemissions, slightly larger than estimates of 8% by Andreset al. [1996] and 5% by Canadell et al. [2007].[20] Air-sea exchange was estimated by several simula-

tions of a 43-box diffusion model including CO2 and 14Cand 13C isotopes [Oeschger et al., 1975], using a pistonvelocity of 14.8 to 18.1 cm hr�1 and an eddy diffusioncoefficient of 3000 to 6000 m2 yr�1. The specification ofpiston velocity uses results from previous studies that con-strained the globally averaged piston velocity with obser-vations of the oceanic inventory of bomb-derived 14C[Sweeney et al., 2007; Naegler et al., 2006] and oceanicD14C and d13C distributions [Krakauer et al., 2006]. Weselected three values across the range in piston velocity(14.8, 16.3 and 18.1 cm hr�1). Then we selected eddy dif-fusion coefficients that allowed the simulated average oce-anic depth profile of natural 14C [Oeschger et al., 1975] andthe simulated oceanic inventory of anthropogenic CO2

[Sabine et al., 2004] to roughly match observations, inaddition to allowing the simulated total bomb-derived 14C inour modeled carbon system (including the atmosphere andterrestrial biosphere, see below) to be consistent withNaegler and Levin [2006]. We used four values for the eddydiffusion coefficient (3000, 4000, 5000 and 6000 m2 yr�1),which are specific to our model setup. One combination ofpiston velocity and eddy diffusion coefficient (18.1 cm hr�1

and 6000 m2 yr�1) did not fit our constraints, so this pair ofparameter values was excluded.[21] Terrestrial ecosystem exchange was estimated by

several simulations of a one-box model of the biosphere.The biosphere was assigned a total preindustrial amount ofoverturning biomass of 470 to 1370 Pg C and a CO2 fertil-ization factor [Keeling et al., 1989] of 0 to 0.4, with globalnet primary production (NPP) of 24 to 42 Pg C yr�1 andaverage ecosystem residence time of 16 to 35 years between1992–2007. Sets of parameter values were chosen to match atotal bomb-derived 14C excess of 615 ⋅ 1026 atoms [Naeglerand Levin, 2006], including the atmospheric inventory andthe ocean inventory from simulations of our one-dimensionalmodel (see above). We allowed a range in total bomb-derived14C of �35 ⋅ 1026 atoms, twice as large as the uncertaintyreported by Naegler and Levin [2006]. We neglect thefraction of NPP for which the assimilated carbon is returnedto the atmosphere within 1–3 years, one-third or more ofNPP [Randerson et al., 2006], since the respiration of such

Figure 3. Schematic of the box model setup and parametervalues used to estimate contributions to the global tropo-spheric D14C trend. Colored lines indicate carbon and isoto-pic fluxes that correspond to the individual contributionsplotted in Figure 4; dashed lines indicate fluxes to strato-spheric reservoirs that affect the troposphere indirectly.


4 of 14

young carbon does not substantially affect troposphericD14C over 1992–2007 (less than 0.4 ‰ yr�1). Allowing forthe fraction of NPP and biomass that are neglected by thisassumption, the total biospheric mass and NPP in our simplemodel (470–1370 Pg C and 24–42 Pg C yr�1) are similarto current dynamic global vegetation models [Cramer et al.,2001; Friedlingstein et al., 2006]. Additional annual carbon,13C and 14C fluxes to or from the biosphere were assignedbased on the land use (LU) flux of Houghton [2008] andthe residual fluxes from a single deconvolution betweenthe fossil fuel and land use sources, the observed atmo-spheric CO2 growth and the box diffusion model represen-tation of oceanic CO2 uptake [Siegenthaler and Oeschger,1987].[22] Cosmogenic production was simulated to occur at an

average rate of 2.16 ⋅ 1026 atoms yr�1, which is 35–40%lower than the rate estimated by Lal [1992] andMasarik andBeer [2009]. The total production rate was reduced from Lal[1992] and Masarik and Beer [2009] in order to match thepre-bomb global 14C inventory simulated by our oceanic andbiospheric models. Observations of 14CO [Manning et al.,2005] support a higher cosmogenic production rate, similarto that predicted by Lal [1992] and Masarik and Beer[2009]. However, as in work by Levin et al. [2010], theuse of a smaller average production rate was necessary toachieve a steady state in our modeled carbon system. Mod-ulation of 14C production by the sunspot cycle was includedaccording to observations of the cosmic neutron flux at Cli-max, Colorado, USA (available at http://ulysses.sr.unh.edu/NeutronMonitor/neutron_mon.html), using relationshipsbetween the cosmic neutron flux and the solar modulationparameter from Lal [1992], Masarik and Beer [1999] andLowe and Allan [2002]. Simulated production varied by�15% [Lal, 1992;Masarik and Beer, 1999] over the sunspotcycle and an uncertainty of �10% in the total productionrate was included. One-half to two-thirds of cosmogenic 14Cproduction was prescribed to occur in the stratosphere, withthe rest occurring in the troposphere [O’Brien, 1979; Jöckelet al., 1999].[23] Production of 14C by nuclear power plants was cal-

culated using energy statistics and emission factors of 14Crelease per unit electrical power generation for 6 differentreactor types [United Nations Scientific Committee on theEffects of Atomic Radiation (UNSCEAR), 2000; Gravenand Gruber, 2011]. Electrical output from each type ofreactor was gathered from the International Atomic EnergyAgency’s Power Reactor Information System (available athttp://www.iaea.org/programmes/a2/). We also accountedfor 14C released by 4 spent nuclear fuel reprocessing facili-ties where 14C emission data was available [UNSCEAR,2000; Schneider and Marignac, 2008; Nakada et al., 2008;UK Environmental Agency, 2008; Anzai et al., 2008], whichamounted to 7–16% of the release from nuclear power plants.Total 14C production from the nuclear energy industrywas 0.28 ⋅ 1026 atoms in 1992, which was 13% of therate of cosmogenic production. Anthropogenic 14C produc-tion increased to 0.36 ⋅ 1026 atoms in 2000 then remainedlargely constant until 2007. Our calculation includes 14Creleased from nuclear power plants as methane since thesereleases will eventually oxidize to form 14CO2. This assumes

that the nuclear-derived atmospheric 14CH4 inventory didnot change substantially, which is reasonable consideringthe increase in production was modest over 1992–2000(30%) and steady thereafter. We assume an uncertainty of�50% in the total production of 14C by the nuclear energyindustry.[24] Finally, the trend in tropospheric D14C caused by

stratosphere-troposphere transport was estimated by a two-box model of the stratosphere with a residence time in thelower box of 1 to 1.5 years and a residence time in the upperbox of 5 years, with 28% of the stratospheric mass in theupper box, similar to Randerson et al. [2002]. Cosmogenicproduction of 14C in the stratosphere was simulated to occureither mainly in the lower stratosphere (75%) or to be dis-tributed equally between the stratospheric boxes.

4.2. Modeled Influences on Tropospheric D14C Trend

[25] Fossil fuel CO2 emissions increased from 6.1 to8.5 Pg C yr�1 between 1992 and 2007, ranging between�0.6 to 5.3% growth each year [Marland et al., 2008;Canadell et al., 2007]. The resulting trend in troposphericD14C was �11.4 � 1.1 ‰ yr�1 in 1992 and decreased to�13.2 � 1.3 ‰ yr�1 in 2007 (Figure 4a and Table 1).The D14C trend caused by fossil fuel emissions changed bya smaller fraction than the emissions themselves becauserising CO2 concentrations and declining D14C in CO2 arereducing the sensitivity of atmospheric D14C to fossil fuel-derived CO2 [Graven et al., 2012; Levin et al., 2010].[26] Air-sea fluxes contributed to a decreasing trend in

D14C and showed the largest change over 1992–2007, from�8.7� 2.0‰ yr�1 in 1992 to �2.4� 1.6‰ yr�1 at the endof 2007 (Figure 4a and Table 1). The disequilibriumbetween mixed layer and tropospheric D14C as predicted bythe box diffusion model shrank from �69 � 22 ‰ in 1992to �21 � 15 ‰ at the end of 2007. In comparison, thesimulated disequilibrium of the preindustrial state was�58� 7‰, which matches observations through our tuningof the model (Section 4.1). The air-sea flux may be consid-ered to consist of a steady state component that roughlybalances cosmogenic production and an anthropogeniccomponent that responds to D14C perturbations fromnuclear sources of 14C and from 14C dilution by fossil fuelemissions. The air-sea disequilibrium is now weaker than thepreindustrial air-sea disequilibrium in our box model simu-lations, suggesting that the total oceanic uptake of 14C mayhave become smaller than cosmogenic production in recentyears. This indicates that the anthropogenic 14C fluxreversed sign. While net removal of 14C from the atmo-sphere continued through 2007, dilution by fossil fuel CO2

may have become a stronger influence on the anthropogenicair-sea 14C flux than bomb-derived excess 14C.[27] The biospheric contribution to the trend in tropo-

spheric D14C was simulated to be relatively constant, 3.7 �1.3‰ yr�1 in 1992 and 4.6 � 1.2‰ yr�1 at the end of 2007(Figure 4a and Table 1). In contrast to the ocean, the dis-equilibrium between the terrestrial biosphere and the tropo-sphere remained relatively constant from 1992 through 2007at +85 � 26 to +99 � 8 ‰. This is because mean D14C inrespired carbon has stopped rising and is now decreasing at


5 of 14

a similar rate as tropospheric CO2 (Section 5.1 and Naeglerand Levin [2009]).[28] In order to match the bomb-14C inventory of 615 �

35 ⋅ 1026 atoms, simulations of the biosphere model with

relatively high NPP were balanced by high biomass, or viceversa, so that fixing one parameter reduced the range ofacceptable values for the other parameters. Therefore, if theuncertainty in one of these parameters was improved, theuncertainty in the other parameters could be tightened usingthe 14C inventory as a constraint.[29] Using the global inventory of bomb-derived 14C as a

constraint also coupled the biospheric parameters to themodeled oceanic inventory. Simulations with slower gasexchange and diffusion reduced the oceanic inventory,requiring longer biospheric residence times to increase thebiospheric inventory. Thus, reduced uncertainty in the oce-anic bomb 14C inventory or better constraints on the rates ofglobal average air-sea gas exchange and on vertical mixingin the oceanic interior would also tighten the range ofacceptable values in the biospheric parameters.[30] Tropospheric production of 14C by cosmogenic and

anthropogenic sources contributed an average of 3.3 �1.3 ‰ yr�1 to the trend of D14C in the global troposphere(Figure 4a). Cosmogenic production in the troposphere,modeled as 33–50% of total cosmogenic production, com-prised an average of 2.5 � 0.8 ‰ yr�1. Solar variabilityassociated with the sunspot cycle enhanced production in1993–99 and 2006–08 and reduced production in 1992 and2000–05 but resulted in only �0.2 ‰ yr�1 variation in theD14C trend. Production of 14C by nuclear power plants andnuclear fuel reprocessing contributed an average of 0.9 �0.5‰ yr�1. Decay of 14C in the troposphere is also includedin the tropospheric production component of Figure 4a,though decay was only �0.1 ‰ yr�1 over the 1992–2007period.[31] According to the 2-box model of the stratosphere, the

transport of 14C-enriched stratospheric air was a positiveinfluence of 5.9 � 1.0 ‰ yr�1 in 1992 that decreased to4.9� 1.0‰ yr�1 at the end of 2007 (Figure 4a and Table 1).The decrease in stratospheric influence was consistent witha slowing in the rate of decrease of tropospheric D14Cbetween 1992 and 1997. After this time D14C in the tropo-sphere decreased at a relatively steady rate and the strato-spheric contribution to the trend remained relativelyconstant. After 1997, approximately 30% of the positiveinfluence from the stratosphere was due to the lag time formixing of tropospheric and stratospheric air and 70% wasdue to cosmogenic production. In the future, the cosmogenicinfluence on D14C in the stratosphere and the troposphere is

Table 1. Simulated Contributions to the Tropospheric D14C Trend (‰ yr�1) From This Work and From Levin et al. [2010] in 1992 andthe End of 2007a

Component

This Work Levin et al. [2010]

1992 2007 Description 1992 2007 Description

Fossil fuel �11.4 �13.2 Marland et al. [2008] �11.7 �13.5 Marland et al. [2007]Ocean �8.7 �2.4 Box diffusion model �8.7 �1.8 Extrapolation of ocean dataBiosphere +3.7 +4.6 1-box model +4.0 +3.5 2-box modelStratosphere (Transport and

Cosmogenic Prod.)+5.9 +4.9 2-box model and scaled neutron

flux data+6.0 +4.8 16-box model and sinusoidal

approx.Troposphere (Nuclear and

Cosmogenic Prod.)+3.1 +3.3 Scaled nuclear power prod.

and scaled neutron flux data+2.8 +3.5 Extrapolation of UNSCEAR [2000]

and sinusoidal approx.Total �7.4 �2.9 �7.5 �3.5

aUncertainties are omitted here for clarity but described in Sections 4.2 and 5.1 and shown in Figure 4 for this work.

Figure 4. (a) Contributions to the global D14C trend bystratospheric exchange (red), biospheric exchange (green),14C production in the troposphere (yellow), oceanicexchange (blue) and fossil fuel combustion (black). Thefilled areas reflect the uncertainty or range of plausiblevalues included in the models for each process and the linesshow the middle of the range. (b) The sum of the modeledcomponents is shown as the gray filled area, where the areaencompasses the sum of the trends in Figure 4a plus andminus a quadrature sum of the range/uncertainty for eachindependent process. The derivative of the seasonallyadjusted observations at La Jolla is shown by the black linein Figure 4b, repeated from Figure 2b.


6 of 14

expected to decline as greater CO2 concentrations will causestronger dilution of cosmogenic 14C.[32] The sum of all contributions to the global D14C trend

is shown by the filled area in Figure 4b. The range ofvalues was calculated by adding together the middle of therange of each contribution (shown as lines in Figure 4a) andcomputing a quadrature sum using one-half of the range ofvalues for each independent process. Simulated ranges in thetrend from fossil fuel combustion, 14C production from thenuclear energy industry and the turnover rate of the strato-spheric box model each made independent contributions touncertainty. The range of values for the sum of the bio-spheric and oceanic components of the trend made a singlecontribution to the uncertainty in the overall trend, since thebiospheric and oceanic parameters of the box models weredetermined in concert. Similarly, as different estimatesshifted cosmogenic 14C production between the stratosphereand troposphere, the range in the sum of cosmogenic pro-duction in the stratosphere and troposphere was also con-sidered to be a single contribution to uncertainty.[33] The global D14C trend predicted by the sum of

components weakened slightly between 1992 and 1996 thenremained fairly steady between 1997 and 2007. The pre-dicted global trend was �7.4 � 2.3 ‰ yr�1 in 1992 and�2.9 � 2.4 ‰ yr�1 at the end of 2007, averaging �4.2 �2.2 ‰ yr�1 over the entire period.[34] The observed, smoothed trend at La Jolla is also

shown in Figure 4b, repeated from Figure 2b. The observedtrend overlaps the modeled trend except when D14Cdecreased more rapidly than average, particularly in 1992–93, 2000, and 2004–05. The average trend inD14C observedat La Jolla, �5.5 � 0.1 ‰ yr�1, lies within the modeledrange of values though it is near to the lower end. Consistentwith the model, observed trends in D14C show a slower rateof decrease in the recent decade (Section 3 and Levin et al.[2010] and Graven et al. [2012]).

5. Discussion

5.1. Tropospheric D14C Trend

[35] The correspondence between the modeled globaltrend and the observed long-term trend at La Jolla suggeststhat the simple formulations we have used to represent theexchanges of carbon and 14C provide a reasonable descrip-tion of recent D14C dynamics. As the observed trend is nearthe lower end of the modeled range, the majority of oursimulations are likely to have overestimated the positivetrend contributions of the biosphere, stratosphere and/orproduction in the troposphere and underestimated the nega-tive trend contributions of the ocean and/or fossil fuels.[36] Simulated trends may be biased by the simplicity of

the models, particularly for the ocean and biosphere. Thebox diffusion model used here does not simulate interme-diate or deep water ventilation in high latitudes by allowingdirect exchange between the atmosphere and sub-surfaceoceanic boxes [Oeschger et al., 1975], unlike some otherbox model formulations [e.g., Siegenthaler, 1983]. Whilethe model parameters were selected to correspond to oceanicbomb 14C and anthropogenic carbon inventories, neglecting3-D transports could overestimate surface D14C in recent

years by excluding high latitude exchange with dense, 14C-depleted water. Additionally, the biospheric enrichmentpredicted by our one-box model is larger than that estimatedby Naegler and Levin [2009] using a two-box model. Thissuggests the release of 14C by terrestrial ecosystems may beoverestimated, although the discrepancy with Naegler andLevin [2009] is reduced by the fact that our one-box modelneglects the fraction of NPP (roughly 1/3) which involvesrapid turnover of assimilated carbon.[37] A similar study of the recent trends in D14C was

conducted by Levin et al. [2010] using a different box modelsetup, summarized in Table 1. In comparison to Levin et al.[2010], our treatment is slightly more complex for air-seaexchange and 14C production and simpler for stratosphericand biospheric exchanges. Levin et al. [2010] extrapolatedsurface D14C from oceanic survey data to estimate air-sea14C fluxes. Their estimated fluxes were similar to the boxdiffusion model in 1992 but changed more rapidly over1992–2007 so that their oceanic contribution to the D14Ctrend was weaker than our box diffusion model in 2007,though both estimates lie within the other estimate’s uncer-tainty. Levin et al. [2010] also extrapolated 14C productionby the nuclear energy industry after 1997 and assumed asinusoidal solar cycle variation in cosmogenic production.As nuclear production increased by only 7% from 1997to 2007, Levin et al.’s [2010] extrapolation is likely tohave overestimated production from nuclear power plantsin recent years. Levin et al.’s [2010] smooth sinusoidalapproximation likely underestimated the sharp fluctuationsin cosmogenic production, so that production was too strongduring solar maxima when cosmogenic production isreduced, particularly the strong solar maximum in 1991–92,and too weak during solar minima when cosmogenic pro-duction is enhanced, though the discrepancy is smaller thanthe total uncertainty. The biospheric 14C flux in Levin et al.[2010] was the same as Naegler and Levin’s [2009] two-box model, as previously mentioned, and simulated a posi-tive contribution that was within 1 ‰ yr�1 of our one-boxmodel. The biospheric contribution decreased slightly inNaegler and Levin’s [2009] model over 1992–2007 whilethe biospheric contribution increased slightly, on average,in our one-box model (Table 1). The different tendenciesreflect a small growth in the troposphere-biosphere disequi-librium in our one-box model versus a small reduction in thetroposphere-biosphere disequilibrium in the two-box modelbetween 1992 and 2007, suggesting the discrepancy betweenthe one-box and two-box models is likely to be larger out-side of the 1992–2007 interval considered here. Positivecontributions of 5–6 ‰ yr�1 were simulated over 1992–2007 by both the 2-box stratosphere model we used andLevin et al.’s [2010] 16-box stratosphere. Both models usedfossil fuel emission data from Marland et al. [2008] leadingto similar contributions to theD14C trend. Differences in thefossil fuel component are likely due to small differences inthe estimated global mean atmospheric composition or thesize of the troposphere.[38] The simulated components of the D14C trend and

the total D14C trend in this work and in the work by Levinet al. [2010] are nearly the same (Table 1), despite thedifferences in model formulation. The average global trend


7 of 14

simulated over 1992–2007 was �4 ‰ yr�1 in our modeland �5 ‰ yr�1 in that of Levin et al. [2010], consistentwithin the uncertainty of �2–3 ‰ yr�1. We note howeverthat both model setups have used Naegler and Levin’s[2006] estimate of total bomb-derived excess 14C as a con-straint on biospheric and oceanic D14C, and that the simu-lated trends have rather large uncertainties (Section 4.2 andLevin et al. [2010]). The average global trend over 1992–2007 is smaller in our model, despite matching that of Levinet al. [2010] in 1992 (Table 1), since Levin et al.’s [2010]simulated trend weakens gradually over the whole periodwhile our simulated trend weakens more rapidly in the firstfew years then remains fairly steady. The largest componentof uncertainty to the trend is in air-sea exchange, suggestingthat additional constraints on the air-sea flux of 14C wouldparticularly improve the uncertainty range of the full mod-eled trend of D14C.

5.2. Rapid Decline of D14C in 2000

[39] The most outstanding feature in the seasonallyadjusted D14C record at La Jolla is the rapid decrease inD14C in 2000 (Figures 2b and 4b). The rate of decreaseappeared to be nearly twice as rapid as the average. Thefeature was also observed at Jungfraujoch, Switzerland butnot at Cape Grim, Australia [Levin et al., 2010], suggestingit extended through the Northern Hemisphere midlatitudesbut not into the Southern Hemisphere. Restriction to theNorthern Hemisphere suggests that anomalous regional 14Cor carbon fluxes or changes in atmospheric circulation innorthern regions may be responsible for the enhanceddecline in D14C.[40] One potential cause of the anomalous decrease in

Northern midlatitude D14C in 2000 may have been astrengthening of the regional air-sea 14C flux in the NorthernPacific Ocean. In 2000–01, winter sea surface temperature inthe Pacific north of 40° was anomalously cold and high windspeeds and exceptionally high gas exchange velocities wereobserved in theWestern North Pacific [Kawabata et al., 2003].

These high wind speeds may have enhanced the ventilationof deep waters in the North Pacific, exposing aged and 14C-poor water masses [Key et al., 2004] and resulting in rapid,anomalous net exchange with lower-D14C CO2 entering theatmosphere and higher-D14C CO2 entering the ocean. Directobservations are not available to verify that changes inD14Coccurred in the surface waters of the North Pacific duringthis time period. However, this process can be investigatedwith observations of atmospheric O2, since anomalous ven-tilation would also enhance oceanic uptake of O2, which isdepleted in aged water masses due to consumption byorganic matter remineralization.[41] Hamme and Keeling [2008] observed a strong

decrease in atmospheric oxygen in the Northern Hemisphereduring 1999–2001 that is concurrent with the strong decreasein our D14C observations at La Jolla. In Figure 5, we showthe trend over 1992–2007 in seasonally adjusted D14C andatmospheric O2/N2 ratios, given as atmospheric potentialoxygen (APO). The APO notation refers to O2/N2 ratiosthat have been corrected for terrestrial biospheric exchange,reported as part per million deviations from a standard ratio[Stephens et al., 1998; Keeling et al., 1998]. We also showthe year-to-year change in annual mean to demonstrate thatthe anomalous decline of bothD14C and APO in 2000 is notlikely to be an artifact of curve fitting procedures. The year-to-year change in annual mean also shows an anomalousdecrease in both D14C and APO, though the anomalies arereduced due to the coarser temporal resolution in the annualmean compared to the spline curves.[42] Hamme and Keeling [2008] estimate that anoma-

lous exposure of deep waters with potential density of1026.6 kg m�3 (the sq 26.6 isopycnal) in the Western NorthPacific could account for the decrease observed in atmo-spheric O2/N2. Using the approximations of Hamme andKeeling [2008] to account for the air-sea O2 flux, we esti-mate that the decrease in D14C could also be explained byunusual exposure of and rapid exchange with cool, deepwaters in the North Pacific. D14C in waters of the sq 26.6isopycnal was observed to be �20 ‰ in 1992 [Key et al.,2004], approximately 30 ‰ lower than D14C in waters ofthe sq 26.4 and 26.5 isopycnals that normally outcrop in theWestern North Pacific near the Kamchatka Peninsula.Assuming that D14C in these isopycnals did not changesubstantially between 1992 and 2000, the exposure of thesq 26.6 isopycnal would have enhanced the local air-seaD14C gradient by 30–40%. The resulting increase in 14Cflux could potentially have caused an anomalous decreaseof �1 to �4.5 ‰ yr�1 in Northern midlatitude air,depending on the spatial extent of anomalous upwelling andthe advection and mixing of the low-D14C air from thePacific. A model study resolving high frequency changes toair-sea fluxes and atmospheric transport from the oceansurface is needed to accurately test this hypothesis.[43] Anomalies in oceanic fluxes occurring in the tropics

are not likely to be responsible for the anomalous decrease inD14C in 2000 or other significant variability in the recenttrend of D14C at La Jolla because the tropical air-sea D14Cgradient observed in the 1990s was weak. Surface waters ofthe equatorial Eastern Pacific during the 1990s had D14C ofapproximately 70 ‰, as measured in dissolved inorganic

Figure 5. The derivative of the seasonally adjusted obser-vations at La Jolla for (top) D14C (repeated from Figure 2b)and (bottom) APO [Hamme and Keeling, 2008], shown asgray lines. The change in annual mean D14C and APO atLa Jolla is shown by the black squares.


8 of 14

carbon by Nydal et al. [1998] and Key et al. [2004] and inshallow corals at Wolf Island by T. Guilderson. This wasonly 30 ‰ lower than average atmospheric D14C in 2000.By contrast, D14C in the sq 26.6 isopycnal water of theNorth Pacific was roughly 125 ‰ below atmospheric D14C.Moreover, the anomalous drop in atmosphericD14C in 2000was not observed at Cape Grim, Australia [Levin et al.,2010], suggesting the cause of the anomaly took place innorthern regions.[44] The rate of decrease in D14C observed at La Jolla in

1992–93 also appeared to be quite rapid, with D14C drop-ping at a similar rate as in 2000. The anomaly in 1993 wasprobably smaller in magnitude than in 2000, however, sincean overall slowing in the rate of decrease of D14C occurredbetween 1993 and 2000. Observations at Jungfraujoch showa similar rate of decrease in 1992–93 as at La Jolla (�10 ‰yr�1), but the trend of D14C in the longer record at Jung-fraujoch clearly slowed between 1986 and 2000 such thatthe trend observed in 1992–93 was not especially prominent[Levin et al., 2010].

5.3. Periodic Variation in D14C

[45] To identify timescales of periodic variability we per-formed a spectral analysis on D14C at La Jolla after adjust-ing the observations to regularly spaced monthly values andlinearly detrending. The monthly values were calculated byfitting a function to the observations that included a lineartrend, an annual harmonic and an error term that was eval-uated using a loose spline (cutoff period of 4 months), andthen evaluating the function at mid-month. A power spec-trum of the D14C observations expressed a strong peak at anannual period; we investigate this seasonal variation in anaccompanying paper [Graven et al., 2012]. A smaller peakwas also present at a period of 28 months.[46] Variation in D14C of atmospheric CO2 at the 28

month (2.3 yr) period has not previously been reported.Variation on periods of 2.6–5.8 yr at Wellington, NewZealand over 1970–1995 was described by Dutta [2002],and related to perturbations in air-sea exchanges [Rozanskiet al., 1995] and, potentially, air-land exchanges causedby the El Niño/Southern Oscillation.[47] The smaller peak at the 28-month period can be

investigated by comparing a spectrum of the D14C data,after removing the seasonal variation, with the correspond-ing red noise spectrum with the same one-lag autocorrelationcoefficient [Wilks, 1995]. The seasonal variation wasremoved from the data in two ways. First, by using the splinecurve described in Section 3. Second, by applying a low-pass Butterworth filter with 10th order and cutoff period of24 months. In both methods, the 28-month period was out-side the 0.05 confidence interval of the corresponding rednoise spectra (Figure 6a), suggesting that observed vari-ability at this period is real and significant. Variation at 2–3 year timescales is also apparent in the observed trend(Figures 2b and 6b) and suggests thatD14C may be sensitiveto climatic variations that operate on similar timescales.[48] One climatic mode that may be associated with D14C

variability of a 28 month period is the Quasi-BiennialOscillation (QBO), a periodic shifting of the zonal winds inthe tropical stratosphere that regulate tropical upwelling intothe stratosphere and the planetary waves that influenceextratropical tropospheric weather patterns and stratosphere-troposphere exchange [Baldwin et al., 2001]. The QBO’sdominant period is also 28 months, the same periodobserved in D14C at La Jolla (Figure 6a). In an atmospherictransport modeling study, Hamilton and Fan [2000] foundthat the QBO had a significant influence on simulatedtropospheric growth rates of long-lived trace gases N2Oand CH4 through its influence on stratosphere-tropospheretransport, suggesting the QBO may also have significanteffects onD14C variability. The effects of the QBO onD14Cvariability are likely to be opposite to the effects on N2O andCH4, since the stratosphere is a source for

14C and a sink forN2O and CH4. Figure 6b shows that positive anomalies inthe trend ofD14C at La Jolla appear to coincide with negativeanomalies in the QBO index at 50 hPa, which correspondto easterly winds and enhanced upwelling in the tropicalstratosphere. The mechanism by which the QBO couldmodulate the tropospheric growth rate of D14C may beassociated with the transport of tropospheric, low-D14C airinto the stratosphere in the tropics, or with the transportof stratospheric, high-D14C air into the troposphere in the

Figure 6. (a) Power spectra of the QBO index and of line-arly detrended, seasonally adjusted D14C at La Jolla usinglow-pass filter and spline techniques as solid lines. Dashedlines show corresponding red noise spectra. The 28-monthperiod is indicated by the vertical gray line. (b) The D14Ctrend anomaly and QBO index for 1992–2007, where theaxis of the QBO index is inverted. The QBO index showsthe zonally averaged 50 hPa zonal wind anomaly at theequator in m s�1 (available at http://www.cpc.ncep.noaa.gov/data/indices/).


9 of 14

extratropics, and may involve substantial lags [Hamilton andFan, 2000]. Studies using an atmospheric model would allowpotential mechanisms of interaction between the QBO andtropospheric D14C to be investigated.

6. Summary

[49] Measurements of D14C in CO2 were conducted onmonthly samples from La Jolla, California, USA collectedbetween July 1992 and December 2007 by the Scripps CO2

Program. D14C analysis was conducted at Lawrence Liver-more National Laboratory by accelerator mass spectrometrywith average measurement uncertainty of �1.9 ‰.[50] The D14C observations fit an average linear trend of

�5.5 � 0.1 ‰ yr�1. The decrease of D14C was highlyvariable, however, and expressed a 28-month periodicity. Astrong decline inD14C in 2000 was concurrent with a strongdecline in atmospheric O2/N2 ratios at La Jolla, suggesting amutual cause from enhanced oceanic ventilation in the NorthPacific [Hamme and Keeling, 2008].[51] Using simple models of the contributions to the global

tropospheric D14C trend, similar to Levin et al. [2010], weshowed that fossil fuel emissions were the strongest influencebetween 1992 and 2007. Oceanic exchange also contrib-uted a negative influence over the entire period 1992–2007; however, the influence weakened by 70%. Themodeled air-sea 14C flux and D14C disequilibrium is nowsmaller than in the preindustrial state, suggesting the per-turbation caused by fossil fuel emissions appears to havebecome more important than the perturbation from nuclearweapons testing. Negative influences from fossil fuelcombustion and oceanic exchange were moderated by 14Cproduction in the troposphere and biospheric and strato-spheric exchange, adding up to an overall rate of changeof �4.1 � 2.2 ‰ yr�1, slower than the average observedtrend but consistent within the uncertainties.

Appendix A: Data Table

[52] Measurements of D14C in CO2 samples collected bythe Scripps CO2 Program and measured at Lawrence Liver-more National Laboratory are provided in Table A1. TheCO2 mole ratio and d13C listed are an average of all mea-surements with the same sample date. The d13C valuesfootnoted with an “a” in Table A1 are estimates of d13C whenmeasurements of d13C in concurrently sampled CO2 werenot available. CO2 mole ratios were measured on the‘SIO 2008A’ Calibration Scale. The SIO calibration scalefor CO2 is established by infrared and manometric analysisof primary reference gases [Keeling et al., 2002]. The SIOcalibration scale is tied to the historic CO2 measurements atSIO and independent of the WMO scale since 1995. d13Cvalues are relative to the international V-PDB standard andinclude the addition of a �0.112 ‰ offset for consistencywith measurements performed at the Center for IsotopeResearch, University of Groningen, Netherlands. sTot is thetotal measurement uncertainty in D14C. Flagged samples(14%) have been removed. D14C measurements at 6 addi-tional SIO clean air sites are reported in the companion paper[Graven et al., 2012].

Table A1. Measurements From La Jolla, California, USA

SIO IDLLNLID

SampleDate

CO2

(ppm)d13C(‰)

D14C(‰)

sTot(‰)

K92-418 116193 01-Jul-92 355.35 �7.815 135.0 1.7K92-487 124125 23-Jul-92 351.31 �7.605 131.5 1.7K92-496 116194 03-Aug-92 352.50 �7.671 133.7 1.7K92-840 116195 09-Sep-92 348.56 �7.514 137.6 1.7K93-004 116196 02-Oct-92 351.67 �7.667 135.2 1.7K93-023 116197 29-Oct-92 354.58 �7.780 132.5 1.7K93-059 116199 11-Dec-92 358.68 �8.046 128.7 1.7K93-081 124395 18-Dec-92 358.76 �8.003 132.3 1.7K93-141 116200 11-Jan-93 359.76 �8.058 129.4 1.7K93-149 116201 09-Feb-93 359.41 �8.039 132.8 1.7K93-208 124402 24-Feb-93 360.03 �8.065 127.7 1.7K93-237 116202 03-Mar-93 360.59 �8.087 129.9 1.7K93-245 116204 12-Apr-93 360.79 �8.027 125.2 1.7K93-356 116205 18-May-93 362.35 �8.106 127.8 1.7K93-431 116206 21-Jun-93 359.62 �7.952 127.3 1.7K93-502 124121 01-Jul-93 357.25 �7.862 120.5 1.7K93-612a 128108 12-Aug-93 351.71 �7.569 123.4 1.7K93-612b 128112 12-Aug-93 351.71 �7.569 124.8 1.7K93-979 131551 11-Nov-93 356.26 �7.810 125.0 1.7K94-107 124385 30-Nov-93 357.98 �7.917 121.4 1.7K94-253a 131029 15-Dec-93 359.93 �8.018 119.6 1.7K94-298 124366 08-Feb-94 360.76 �8.034 116.1 1.7K94-355 131095 11-Mar-94 360.94 �8.024 119.7 1.7K94-402 124122 04-Apr-94 362.71 �8.174 111.8 1.7K94-675 124123 17-May-94 363.79 �8.222 115.2 1.7K94-729 124126 22-Jun-94 359.73 �7.987 115.1 1.7K94-952 124128 24-Aug-94 352.63 �7.627 114.9 1.7K94-951 116209 24-Aug-94 352.63 �7.627 118.5 1.7K95-006 124136 02-Nov-94 358.39 �7.890 118.7 1.7K95-018 116212 18-Nov-94 360.65 �8.014 118.7 1.7K95-179 116215 03-Jan-95 361.47 �8.066 119.4 1.7K95-581 116216 08-Feb-95 362.79 �8.150 116.8 1.7K95-587 124137 15-Feb-95 362.54 �8.186 110.3 1.7K95-724 116217 13-Mar-95 363.59 �8.181 113.4 1.7K95-755 116218 07-Apr-95 363.60 �8.141 109.0 1.7K95-846 116219 15-May-95 365.56 �8.247 114.5 1.7K95-851 101933 08-Jun-95 363.99 �8.275 109.5 2.3K95-850 104335 08-Jun-95 363.99 �8.275 112.4 2.3K95-852 116220 08-Jun-95 363.99 �8.275 113.8 1.7K95-A48 104336 03-Jul-95 360.79 �7.966 112.4 2.4K95-A49 101934 03-Jul-95 360.79 �7.966 112.9 2.2K95-C89 124160 21-Aug-95 353.10 �7.578 107.9 1.7K95-C87 101935 21-Aug-95 353.10 �7.578 110.2 2.2K95-C86 104337 21-Aug-95 353.10 �7.578 110.2 2.3K95-E07 104338 26-Sep-95 357.76 �7.809 106.5 2.4K95-E08 101936 26-Sep-95 357.76 �7.809 108.3 2.2K96-023 101937 16-Oct-95 358.38 �7.790 107.9 2.3K96-022 104339 16-Oct-95 358.38 �7.790 112.5 2.3K96-035 101938 10-Nov-95 360.49 �7.939 108.5 2.2K96-034 104340 10-Nov-95 360.49 �7.939 109.7 2.3K96-132 103195 22-Jan-96 364.56 �8.153 105.0 2.2K96-131 103194 22-Jan-96 364.56 �8.153 106.8 2.2K96-149a 124406 02-Feb-96 365.31 �8.154 112.8 1.7K96-149b 124407 02-Feb-96 365.31 �8.154 116.1 1.7K96-185 103196 22-Feb-96 365.91 �8.224 103.9 2.3K96-187 124124 22-Feb-96 365.91 �8.224 104.1 1.7K96-186 103197 22-Feb-96 365.91 �8.224 104.2 2.2K96-297 103199 11-Mar-96 365.16 �8.171 104.6 2.1K96-296 103198 11-Mar-96 365.16 �8.171 107.8 2.2K96-321 124158 01-Apr-96 366.10 �8.217 103.0 1.7K96-378 103201 16-Apr-96 366.48 �8.222 102.5 2.2K96-377 103200 16-Apr-96 366.48 �8.222 105.2 2.2K96-450 103202 03-May-96 367.60 �8.204 102.2 2.2K96-451 103203 03-May-96 367.60 �8.204 106.1 2.2K96-518 103204 26-Jun-96 363.05 �8.027 104.1 2.2K96-519 104346 26-Jun-96 363.05 �8.027 104.3 2.3K96-711 124162 05-Aug-96 358.46 �7.775 104.0 1.7K96-837 124381 04-Sep-96 356.25 �7.680 105.4 1.7K96-921 131538 25-Oct-96 361.37 �7.942 108.1 1.7K96-938 124171 27-Nov-96 364.09 �8.075 104.8 1.7K97-112 141146 03-Jan-97 364.45 �8.137 101.5 2.2


10 of 14

Table A1. (continued)

SIO IDLLNLID

SampleDate

CO2

(ppm)d13C(‰)

D14C(‰)

sTot(‰)

K97-150 124172 17-Feb-97 365.88 �8.161 98.9 1.7K97-151 124380 17-Feb-97 365.88 �8.161 101.4 1.7K97-213 131110 22-Apr-97 367.81 �8.170 100.2 1.7K97-375 124174 23-Jun-97 364.68 �8.053 100.7 1.7K97-381 124400 30-Jun-97 363.74 �8.022 103.4 1.7K97-465 124178 18-Jul-97 359.32 �7.752 99.2 1.7K97-466 138143 18-Jul-97 359.32 �7.752 103.9 2.2K97-569 131570 10-Aug-97 357.23 �7.634 102.5 1.7K97-660b 128095 10-Oct-97 360.19 �7.792 101.3 1.7K97-660a 128074 10-Oct-97 360.19 �7.792 101.7 1.7A98-005 124177 10-Jan-98 365.99 �8.092 101.3 1.7A98-011 124367 13-Jan-98 367.52 �8.180 100.3 1.7K98-102b 126937 13-Feb-98 367.29 �8.150 97.4 1.7K98-102a 126910 13-Feb-98 367.29 �8.150 99.0 1.7A98-183 125613 23-Apr-98 370.04 �8.297 96.6 1.7A98-197 124212 22-May-98 370.58 �8.312 92.5 1.7A98-198 124388 22-May-98 370.58 �8.312 93.3 1.7A98-272 101940 29-Jun-98 364.02 �7.958 96.6 2.2A98-474 125579 29-Jul-98 363.43 �7.884 95.1 1.7A98-473 101942 29-Jul-98 363.43 �7.884 98.9 2.2A98-472 101941 29-Jul-98 363.43 �7.884 99.3 2.2A98-480 101943 25-Aug-98 360.37 �7.759 96.8 2.6A98-481 101944 25-Aug-98 360.37 �7.759 98.7 2.3A99-040 101945 29-Oct-98 366.91 �8.094 97.1 2.2K99-029 101946 29-Oct-98 366.91 �8.094 97.9 2.2K99-035 101948 14-Dec-98 370.09 �8.264 95.1 2.5A99-046 101947 14-Dec-98 370.09 �8.264 97.9 2.2K99-044 101949 26-Jan-99 370.21 �8.268 90.5 2.3A99-130 101950 26-Jan-99 370.21 �8.268 92.3 2.2K99-050 138125 08-Feb-99 371.34 �8.288 95.1 2.2A99-243 124180 12-Apr-99 372.90 �8.353 93.6 1.7A99-511 125603 21-Jul-99 363.01 �7.842 91.8 1.7A99-512 124208 21-Jul-99 363.01 �7.842 93.8 1.7A99-551 125614 10-Aug-99 364.68 �7.881 90.9 1.7A99-552 125624 10-Aug-99 364.68 �7.881 93.3 1.7A99-760 124404 08-Sep-99 361.55 �7.751 94.9 1.7A99-766 126967 15-Oct-99 366.34 �7.979 94.6 1.7A00-009 124176 16-Nov-99 367.85 �8.044 95.1 1.7A00-014 124371 17-Nov-99 369.03 �8.125 97.0 1.7A00-122 125588 31-Dec-99 369.80 �8.130 92.8 1.7A00-128 127000 21-Jan-00 370.28 �8.129 94.2 1.7A00-185 125590 11-Feb-00 372.05 �8.267 91.6 1.7A00-210 141141 22-Feb-00 372.56 �8.285 85.0 2.2A00-289 138077 20-Mar-00 374.38 �8.348 83.3 2.2A00-288 124386 20-Mar-00 374.38 �8.348 86.0 1.7A00-304 124205 14-Apr-00 373.79 �8.335 85.1 1.7A00-412 124209 26-May-00 374.37 �8.348 83.5 1.7A00-419 124210 05-Jun-00 372.08 �8.225 83.7 1.7A00-448 104352 16-Jun-00 371.90 �8.208 81.2 2.2A00-447 103210 16-Jun-00 371.90 �8.208 81.8 2.2A00-567 103211 14-Jul-00 365.91 �7.886 85.0 2.2A00-571 124213 14-Jul-00 365.91 �7.886 85.4 1.7A00-568 104353 14-Jul-00 365.91 �7.886 85.8 2.3A00-607 104354 14-Aug-00 362.61 �7.733 82.6 2.2A00-606 103212 14-Aug-00 362.61 �7.733 83.3 2.2A00-609 124214 14-Aug-00 362.61 �7.733 83.5 1.7A00-615 124175 18-Aug-00 360.34 �7.603 87.4 1.7A00-718 104355 05-Sep-00 362.31 �7.733 82.9 2.2A00-719 103213 05-Sep-00 362.31 �7.733 85.8 2.2A00-730 104356 10-Oct-00 367.34 �7.960 84.9 2.2A00-731 103214 10-Oct-00 367.34 �7.960 85.5 2.2A01-085 103215 09-Nov-00 370.23 �8.106 83.1 2.3A01-086 104357 09-Nov-00 370.23 �8.106 84.3 2.1A01-121 104366 08-Jan-01 372.10 �8.156 83.5 2.2A01-122 104367 08-Jan-01 372.10 �8.156 84.9 2.3A01-128 104369 07-Feb-01 373.84 �8.244 79.5 2.1A01-127 104368 07-Feb-01 373.84 �8.244 80.0 2.1A01-188 104371 07-Mar-01 375.11 �8.328 75.2 2.2A01-187 104370 07-Mar-01 375.11 �8.328 75.9 2.4A01-217 124364 23-Mar-01 373.77 �8.256 77.8 1.7A01-233 104373 02-Apr-01 375.87 �8.363 73.2 2.1


SIO IDLLNLID

SampleDate

CO2

(ppm)d13C(‰)

D14C(‰)

sTot(‰)

A01-232 104372 02-Apr-01 375.87 �8.363 75.4 2.3A01-308 104374 04-May-01 376.70 �8.422 73.2 2.2A01-309 104375 04-May-01 376.70 �8.422 74.2 2.2A01-302 104376 04-Jun-01 375.73 �8.326 76.2 2.1A01-303 104377 04-Jun-01 375.73 �8.326 76.2 2.3A01-298 128062 04-Jun-01 375.73 �8.326 78.2 1.7A01-358 128093 13-Jun-01 373.27 �8.212 76.8 1.7A01-363 104342 13-Jun-01 373.27 �8.212 79.4 2.4A01-362 104341 13-Jun-01 373.27 �8.212 80.0 2.3A01-383 104343 16-Jul-01 366.73 �7.860 82.5 2.2A01-382 104334 16-Jul-01 366.73 �7.860 84.6 2.3A01-574 124217 24-Jul-01 364.55 �7.750 86.0 1.7A01-584 104344 10-Aug-01 365.22 �7.792 80.9 2.2A01-585 104345 10-Aug-01 365.22 �7.792 84.1 2.3A01-599 104381 06-Sep-01 363.82 �7.739 83.3 2.2A01-598 104380 06-Sep-01 363.82 �7.739 83.8 2.2A01-613 104383 31-Oct-01 370.02 �8.011 75.9 2.2A01-609 141173 31-Oct-01 370.02 �8.011 77.0 2.2A01-612 104382 31-Oct-01 370.02 �8.011 78.5 2.2A02-092 104384 09-Dec-01 372.87 �8.185 73.5 2.2A02-093 104385 09-Dec-01 372.87 �8.185 75.6 2.2A02-170 101951 09-Jan-02 374.34 �8.216 74.6 2.2A02-165 128065 09-Jan-02 374.34 �8.216 76.9 1.7A02-169 104386 09-Jan-02 374.34 �8.216 80.1 2.5A02-186 101952 17-Feb-02 375.21 �8.274 74.0 2.2A02-185 104387 17-Feb-02 375.21 �8.274 75.1 2.8A02-225 101953 22-Mar-02 376.98 �8.365 70.9 2.2A02-224 104388 22-Mar-02 376.98 �8.365 73.6 2.5A02-221 124218 22-Mar-02 376.98 �8.365 77.1 1.7A02-254 104389 25-Apr-02 376.87 �8.334 71.5 2.6A02-255 101954 25-Apr-02 376.87 �8.334 77.3 2.2A02-353 103192 13-May-02 377.25 �8.332 68.1 2.2A02-352 101931 13-May-02 377.25 �8.332 68.5 2.3A02-368 124195 10-Jun-02 375.99 �8.265 75.0 1.7A02-419 101932 21-Jun-02 373.09 �8.123 68.4 2.4A02-420 103193 21-Jun-02 373.09 �8.123 69.3 2.2A02-465 117787 12-Jul-02 371.13 �7.998 72.6 1.7A02-469 124215 24-Jul-02 368.98 �7.892 74.3 1.7A02-470 125591 24-Jul-02 368.98 �7.892 79.2 1.7A02-572 117796 06-Aug-02 364.89 �7.713 74.1 1.7A02-584 117802 09-Sep-02 367.55 �7.858 70.4 1.7A02-766 125584 11-Oct-02 371.31 �8.021 72.7 1.7A02-765 117842 11-Oct-02 371.31 �8.021 72.7 2.9A03-058 117847 09-Nov-02 372.65 �8.063 70.2 2.7A03-149 117846 17-Dec-02 375.99 �8.284 70.8 2.8A03-155 128142 29-Dec-02 376.84 �8.303 69.5 1.7A03-162 126976 07-Feb-03 378.07 �8.377 66.6 1.7A03-161 117794 07-Feb-03 378.07 �8.377 75.4 1.7A03-168 141154 14-Feb-03 378.03 �8.339 71.6 2.2A03-259 128087 04-Mar-03 377.14 �8.279 71.1 1.7A03-265 141197 18-Mar-03 379.18 �8.415 67.7 2.2A03-275 125585 10-Apr-03 379.49 �8.405 65.3 1.7A03-274 117839 10-Apr-03 379.49 �8.405 66.2 2.7A03-426 125593 24-Apr-03 380.88 �8.508 71.8 1.7A03-425 124403 24-Apr-03 380.88 �8.508 76.2 1.7A03-460 128119 04-Jun-03 380.03 �8.409 69.1 1.7A03-473 131036 24-Jun-03 376.67 �8.252 67.7 1.7A03-472 138106 24-Jun-03 376.67 �8.252 70.4 2.2A03-601 117794 29-Aug-03 367.46 �7.787 68.4 1.7A03-608 125623 09-Sep-03 367.95 �7.772 71.1 1.7A03-607 125602 09-Sep-03 367.95 �7.772 73.1 1.7A04-013 126904 31-Oct-03 375.35 �8.156 69.7 1.7A04-020 128129 16-Nov-03 377.45 �8.285 65.8 1.7A04-019 126960 16-Nov-03 377.45 �8.285 67.4 1.7A04-117 126922 26-Dec-03 379.21 �8.347 67.9 1.7A04-247 124409 31-Jan-04 380.44 �8.406 69.6 1.7A04-263 126943 26-Feb-04 380.52 �8.408 63.3 1.7A04-269 124378 24-Mar-04 381.56 �8.426 61.8 1.7A04-276 125605 26-Mar-04 380.97 �8.403 65.1 1.7A04-275 125604 26-Mar-04 380.97 �8.403 66.1 1.7A04-424 125608 20-Apr-04 382.02 �8.476 62.5 1.7


11 of 14

[53] Acknowledgments. The air sampling and CO2 extractions weresupported by a grant from BP, by the National Science Foundation (NSF)under grant ATM-0632770, the U.S. Department of Energy (DOE) undergrant DE-FG02-07ER64362 as well as by previous awards from NSF andDOE. This work was performed in part under the auspices of the U.S.Department of Energy by Lawrence Livermore National Laboratory undercontract W-7405-Eng-48 and DE-AC52-07NA27344. Radiocarbon analy-ses were funded by grants from NOAA’s Office of Global Programs(NA05OAR4311166) and LLNL’s Directed Research and Developmentprogram (06-ERD-031) to T.P.G. H.D.G. received support from the UCOffice of the President and a NASA ESS Fellowship. H.D.G. also thanksNicolas Gruber for support and helpful discussions. Alane Bollenbacherconducted CO2 and stable isotope analyses. We thank the anonymousreviewers and Jocelyn Turnbull for helpful comments, Ingeborg Levin forsharing recent D14C data from Jungfraujoch and Christian Rödenbeck forassistance with the TM3 Model. This research was also presented in H.D.G.’s doctoral dissertation at the University of California, San Diego,USA, 2008.

ReferencesAndres, R. J., G. Marland, I. Fung, and E. Matthews (1996), A 1° � 1°distribution of carbon dioxide emissions from fossil fuel consumptionand cement manufacture, 1950–1990, Global Biogeochem. Cycles, 10,419–429.

Anzai, K., S. Keta, M. Kano, N. Ishihara, T. Moriyama, Y. Okamura,K. Ogaki, and K. Nodaa (2008), Radioactive effluent releases fromRokkasho Reprocessing plant 1. Gaseous effluent, technical report,Reprocess. Bus. Div., Japan Nuclear Fuel Limited, Rokkasho, Japan.

Baldwin, M. P., et al. (2001), The quasi-biennial oscillation, Rev. Geophys.,39(2), 179–230.

Canadell, J., et al. (2007), Contributions to accelerating atmospheric CO2growth from economic activity, carbon intensity, and efficiency of naturalsinks, Proc. Natl. Acad. Sci. U. S. A., 104(47), 18,866–18,870.

Cantrell, C. A. (2008), Technical Note: Review of methods for linear least-squares fitting of data and application to atmospheric chemistry problems,Atmos. Chem. Phys., 8, 5477–5487.

Conil, S., and A. Hall (2006), Local regimes of atmospheric variability:A case study of Southern California, J. Clim., 19, 4308–4325.

Conway, T. J., and P. Tans (2004), Atmospheric carbon dioxide mixingratios from the NOAA CMDL Carbon Cycle Cooperative Global AirSampling Network (2004), in Trends: A Compendium of Data on GlobalChange, ftp://cdiac.ornl.gov/pub/ndp005/README_co2.html, CarbonDioxide Inf. Anal. Cent., Oak Ridge Natl. Lab., U.S. Dep. of Energy,Oak Ridge, Tenn.

Cramer, W., et al. (2001), Global response of terrestrial ecosystem structureand function to CO2 and climate change: Results from six dynamic globalvegetation models, Global Change Biol., 7, 357–373.

Currie, K. I., G. Brailsford, S. Nichol, A. Gomez, R. Sparks, K. R. Lassey,and K. Riedel (2009), Tropospheric 14CO2 atWellington, NewZealand: Theworld’s longest record, Biogeochemistry, 104(1–3), 5–22, doi:10.1007/s10533-009-9352-6.

Dutta, K. (2002), Coherence of tropospheric 14CO2 with El Niño/SouthernOscillation,Geophys. Res. Lett., 29(20), 1987, doi:10.1029/2002GL014753.

Enting, I. G. (1987), On the use of smoothing splines to filter CO2 data,J. Geophys. Res., 92(9), 10,977–10,984.

Fallon, S. J., T. P. Guilderson, and T. A. Brown (2007), CAMS/LLNL ionsource efficiency revisited, Nucl. Inst. Phys. Res. B, 259, 106–110.

Friedli, H., H. Lötscher, H. Oeschger, U. Siegenthaler, and B. Stauffer(1986), Ice core record of 13C/12C ratio of atmospheric CO2 in the pasttwo centuries, Nature, 324, 237–238.

Friedlingstein, P., et al. (2006), Climate-carbon cycle feedback analysis:Results from the C4MIP model intercomparison, J. Clim., 19, 3337–3353.

Goudriaan, J. (1992), Biosphere structure, carbon sequestering potential andthe atmospheric 14C carbon record, J. Exper. Bot., 43(8), 1111–1119.

Graven, H. D. (2008), Advancing the use of radiocarbon in studies of globaland regional carbon cycling with high precision measurements of 14C inCO2 from the Scripps CO2 Program, Ph.D. thesis, Scripps Inst. of Ocea-nogr., Univ. of Calif., San Diego, La Jolla.

Graven, H. D., and N. Gruber (2011), Continental-scale enrichment ofatmospheric 14CO2 from the nuclear power industry: Potential impacton the estimation of fossil fuel-derived CO2, Atmos. Chem. Phys., 11,12,339–12,349, doi:10.5194/acp-11-12339-2011.

Graven, H. D., T. P. Guilderson, and R. F. Keeling (2007), Methods forhigh-precision 14C AMS measurement of atmospheric CO2 at LLNL,Radiocarbon, 49, 349–356.

Graven, H. D., B. B. Stephens, T. P. Guilderson, T. L. Campos, D. S.Schimel, J. E. Campbell, and R. F. Keeling (2009), Vertical profiles


SIO IDLLNLID

SampleDate

CO2

(ppm)d13C(‰)

D14C(‰)

sTot(‰)

A04-447 125617 26-May-04 379.87 �8.353 63.8 1.7A04-530 116149 09-Jun-04 380.44 �8.369 61.9 1.7A04-529 116148 09-Jun-04 380.44 �8.369 63.7 1.8A04-532 124370 09-Jun-04 380.44 �8.369 66.7 1.7A04-691 117822 27-Jul-04 374.98 �8.039 64.3 2.7A04-690 117808 27-Jul-04 374.98 �8.039 66.6 1.7A04-735 117865 28-Jul-04 373.33 �7.994 60.8 1.7A04-734 117823 28-Jul-04 373.33 �7.994 62.1 3.3A04-740 126932 26-Aug-04 369.73 �7.782 66.4 1.7A04-786 126990 14-Sep-04 370.29 �7.819 66.2 1.7A04-792 125569 10-Oct-04 375.30 �8.066 60.1 1.7A04-793 125587 10-Oct-04 375.30 �8.066 62.9 1.7A05-047 128155 21-Dec-04 380.36 �8.272 65.7 1.7AORG416 113089 26-Jan-05 380.05 �8.237 58.1 1.7CDRG249 113085 26-Jan-05 380.05 �8.237 59.4 1.7CDRG248 113084 26-Jan-05 380.05 �8.237 60.2 1.7CDRG250 113086 26-Jan-05 380.05 �8.237 60.6 1.7AORG420 113090 26-Jan-05 380.05 �8.237 60.7 1.7AORG338 113088 26-Jan-05 380.05 �8.237 61.4 1.7AORG126 113087 26-Jan-05 380.05 �8.237 61.9 1.7CDRG247 113083 26-Jan-05 380.05 �8.237 63.3 1.7A05-236 125622 07-Feb-05 382.09 �8.376 60.3 1.7AORG262 116168 23-Mar-05 383.41 �8.450 58.6 2.1CDRG222 116171 23-Mar-05 383.41 �8.450 58.9 1.7CDRG219 116170 23-Mar-05 383.41 �8.450 60.5 2.2AORG366 116145 07-Apr-05 383.95 �8.510 54.8 1.7A05-288 116166 07-Apr-05 383.95 �8.510 55.6 2.2A05-308 125581 19-Apr-05 384.29 �8.506 51.4 1.7A05-309 125566 19-Apr-05 384.29 �8.506 51.8 1.7A05-351 125567 03-May-05 384.02 �8.465 50.5 1.7A05-349 124394 03-May-05 384.02 �8.465 57.2 1.7A05-472 125610 18-Jun-05 382.18 �8.375 52.4 1.7A05-470 125568 18-Jun-05 382.18 �8.375 53.0 1.7A05-537 141119 05-Jul-05 378.64 �8.197 58.7 2.2A05-561 124377 28-Jul-05 374.95 �8.004 54.9 1.7A05-560 124373 28-Jul-05 374.95 �8.004 59.2 1.7A05-680 125570 07-Sep-05 373.02 �7.917 56.2 1.7A05-681 125570 07-Sep-05 373.02 �7.917 59.7 1.7A05-721 125592 27-Oct-05 378.91 �8.216 59.7 1.7A05-722 125609 27-Oct-05 378.91 �8.216 61.4 1.7A05-777 124383 03-Nov-05 378.48 �8.143 57.6 1.7A05-778 124401 03-Nov-05 378.48 �8.143 59.5 1.7A06-177 131546 25-Jan-06 384.43 �8.388 53.5 1.7A06-230 131038 15-Feb-06 385.30 �8.445 53.1 1.7A06-244 131502 17-Mar-06 386.10 �8.503 50.4 1.7A06-248 131105 17-Mar-06 386.10 �8.503 51.2 1.7A06-315 131123 21-Apr-06 386.41 �8.491 50.7 1.7A06-317 141188 21-Apr-06 386.41 �8.491 56.6 2.2A06-381 131085 22-May-06 387.10 �8.535 58.6 1.7A06-412 131542 12-Jun-06 383.75 �8.357 56.0 1.7A06-543 131519 17-Jul-06 381.26 �8.194 56.3 1.7A06-592 131077 03-Aug-06 377.03 �7.996 57.3 1.7A06-593 131053 13-Sep-06 376.65 �7.987 54.9 1.7A06-632 131137 06-Oct-06 378.45 �8.061 56.3 1.7A06-633 138036 06-Oct-06 378.45 �8.061 57.6 2.2A06-641 131511 09-Nov-06 383.20 �8.274 56.0 1.7A07-087 138130 11-Jan-07 386.06 �8.444 52.0 2.2A07-128 141192 23-Feb-07 386.98 �8.461 51.4 2.2A07-127 138101 23-Feb-07 386.98 �8.461 53.9 2.2A07-198 138075 27-Mar-07 387.85 �8.50a 46.7 2.2A07-330 138062 04-May-07 388.27 �8.50a 48.7 2.2A07-404 138092 07-Jun-07 388.06 �8.50a 44.1 2.2A07-572 138112 17-Jul-07 380.83 �8.20a 55.4 2.2A07-581 138104 07-Aug-07 377.49 �8.00a 52.0 2.2A07-599 138134 06-Sep-07 377.51 �8.00a 51.9 2.2A08-055 141128 12-Oct-07 381.60 �8.10a 53.3 2.2A08-061 141136 07-Dec-07 385.72 �8.30a 42.3 2.2

aEstimated d13C values, when direct measurements were not available.


12 of 14

of biospheric and fossil fuel-derived CO2 and fossil fuel CO2:CO ratiosfrom airborne measurements of D14C, CO2 and CO above Colorado,USA, Tellus, Ser. B, 61, 536–546.

Graven, H. D., T. P. Guilderson, and R. F. Keeling (2012), Observations ofradiocarbon in CO2 at seven global sampling sites in the Scripps flasknetwork: Analysis of spatial gradients and seasonal cycles, J. Geophys.Res., 117, D02302, doi:10.1029/2011JD016535.

Guenther, P. R., A. F. Bollenbacher, C. D. Keeling, E. F. Stewart, andM. Wahlen (2001), Calibration methodology for the Scripps 13C/12Cand 18O/16O stable isotope program, 1996–2000, report, 118 pp.,Scripps Inst. of Oceanogr. Univ. of Calif., San Diego, La Jolla.

Hamilton, K., and S. M. Fan (2000), Effects of the stratospheric quasi-biennial oscillation on long-lived greenhouse gases in the troposphere,J. Geophys. Res., 105(D16), 20,581–20,588.

Hamme, R. C., and R. F. Keeling (2008), Ocean ventilation as a driver ofinterannual variability in atmospheric potential oxygen, Tellus, Ser. B,60(5), 706–717.

Hesshaimer, V., M. Heimann, and I. Levin (1994), Radiocarbon evidencefor a smaller oceanic carbon dioxide sink than previously believed,Nature, 370(6486), 201–203.

Houghton, R. (2008), Carbon flux to the atmosphere from land-usechanges: 1850–2005, in Trends: A Compendium of Data on GlobalChange, http://cdiac.ornl.gov/trends/landuse/houghton/houghton.html,Carbon Dioxide Inf. Anal. Cent., Oak Ridge Natl. Lab., U.S. Dep. ofEnergy, Oak Ridge, Tenn.

Hsueh, D. Y., N. Y. Krakauer, J. T. Randerson, X. Xu, S. E. Trumbore, andJ. R. Southon (2007), Regional patterns of radiocarbon and fossil fuel-derived CO2 in surface air across North America, Geophys. Res. Lett.,34, L02816, doi:10.1029/2006GL027032.

Jöckel, P., M. G. Lawrence, and C. A. M. Brenninkmeijer (1999), Simula-tions of cosmogenic 14CO using the three-dimensional atmosphericmodel MATCH: Effects of 14C production distribution and the solarcycle, J. Geophys. Res., 104(D9), 11,733–11,743.

Kawabata, H., H. Narita, K. Harada, S. Tsunogai, and M. Kusakabe (2003),Air-sea gas transfer velocity in stormy winter estimated from radon defi-ciency, J. Oceanogr., 59(5), 651–661.

Keeling, C. D. (1979), The Suess effect: 13Carbon-14Carbon interrelations,Environ. Int., 2(4–6), 229–300.

Keeling, C. D., and S. C. Piper (2001), Exchanges of atmospheric CO2 and13CO2 with the terrestrial biosphere and oceans from 1978 to 2000. IV.Critical overview, SIO Ref. Ser. 01–09, 23 pp., Scripps Inst. of Oceanogr.,Univ. of Calif., San Diego, La Jolla.

Keeling, C. D., and T. P. Whorf (2005), Atmospheric CO2 records fromsites in the SIO air sampling network, in Trends: A Compendium of Dataon Global Change, Carbon Dioxide Inf. Anal. Cent., Oak Ridge Natl.Lab., U.S. Dep. of Energy, Oak Ridge, Tenn., http://cdiac.ornl.gov/trends/co2/sio-keel.html.

Keeling, C. D., R. B. Bacastow, A. F. Carter, S. C. Piper, T. P. Whorf,M. Heimann, W. G. Mook, and H. Roeloffzen (1989), A three-dimen-sional model of atmospheric CO2 transport based on observed winds: 1.Analysis of observational data, in Aspects of Climate Variability in thePacific and the Western Americas, Geophys. Monogr. Ser., vol. 55, editedby D. H. Peterson, pp. 165–235, AGU, Washington, D. C.

Keeling, C. D., P. R. Guenther, G. Emanuele, A. Bollenbacher, andD. J. Moss (2002), Scripps reference gas calibration system for carbondioxide-in-nitrogen and carbon dioxide-in-air standards: Revision of1999, a report prepared for the Global Environmental Monitoring Pro-gram of the World Meteorological Organization, 83 pp., World Meteorol.Org., Geneva.

Keeling, C. D., A. F. Bollenbacher, and T. P. Whorf (2005), Monthly atmo-spheric 13C/12C isotopic ratios for 10 SIO stations, in Trends: A Com-pendium of Data on Global Change, http://cdiac.esd.ornl.gov/trends/co2/iso-sio/iso-sio.html, Carbon Dioxide Inf. Anal. Cent., Oak RidgeNatl. Lab., U.S. Dep. of Energy, Oak Ridge, Tenn.

Keeling, C. D., S. C. Piper, T. P. Whorf, and R. F. Keeling (2011), Evolu-tion of natural and anthropogenic fluxes of atmospheric CO2 from 1957to 2003, Tellus, Ser. B, 63, 1–22.

Keeling, R. F., A. C. Manning, E. M. McEvoy, and S. R. Shertz (1998),Methods for measuring changes in atmospheric O2 concentration andtheir application in southern hemisphere air, J. Geophys. Res., 103(D3),3381–3397.

Key, R. M., et al. (2004), A global ocean carbon climatology: Results fromGlobal Data Analysis Project (GLODAP), Global Biogeochem. Cycles,18, GB4031, doi:10.1029/2004GB002247.

Krakauer, N., J. T. Randerson, F. W. Primeau, N. Gruber, and D. Menemenlis(2006), Carbon isotope evidence for the latitudinal distribution and windspeed dependence of the air-sea gas transfer velocity, Tellus, Ser. B, 58(5),390–417.

Lal, D. (1992), Expected secular variation in the global terrestrial produc-tion rate of radiocarbon, in The Last Deglaciation: Absolute andRadiocarbon Chronologies, edited by E. Bard and W. S. Broecker,pp. 113–126, Springer, Berlin.

Lal, D., and Rama (1966), Characteristics of global tropospheric mixingbased on man-made 14C, 3H, and 90Sr, J. Geophys. Res., 71, 2865–2874.

Levin, I., and V. Hesshaimer (2000), Radiocarbon—A unique tracer ofglobal carbon cycle dynamics, Radiocarbon, 42(1), 69–80.

Levin, I., and B. Kromer (2004), The tropospheric 14CO2 level in mid-latitudes of the Northern Hemisphere (1959–2003), Radiocarbon, 46(3),1261–1272.

Levin, I., and C. Rödenbeck (2008), Can the envisaged reductions of fossilfuel CO2 emissions be detected by atmospheric observations?, Naturwis-senschaften, 95(3), 203–208.

Levin, I., B. Kromer, H. Schoch-Fischer, M. Bruns, M. Münnich, B. Berdau,J. C. Vogel, and K. O. Münnich (1985), 25 years of tropospheric 14Cobservations in central Europe, Radiocarbon, 27(1), 1–19.

Levin, I., J. Schuchard, B. Kromer, and K. O. Münnich (1989), The conti-nental European Suess Effect, Radiocarbon, 31(3), 431–440.

Levin, I., B. Kromer, M. Schmidt, and H. Sartorius (2003), A novelapproach for independent budgeting of fossil fuel CO2 over Europe by14CO2 observations, Geophys. Res. Lett., 30(23), 2194, doi:10.1029/2003GL018477.

Levin, I., B. Kromer, L. P. Steele, and L. W. Porter (2007), Continuousmeasurements of 14C in atmospheric CO2 at Cape Grim, 1997–2006, inBaseline Atmospheric Program Australia 2005–2006, pp. 57–59, Aust.Bur. of Meteorol., Melbourne, Victoria, Australia.

Levin, I., et al. (2010), Observations and modelling of the global distribu-tion and long-term trend of atmospheric 14CO2, Tellus, Ser. B, 62, 26–46.

Lowe, D. C., and W. Allan (2002), A simple procedure for evaluatingglobal cosmogenic 14C production in the atmosphere using neutron mon-itor data, Radiocarbon, 44(1), 149–157.

Manning, M. R., D. C. Lowe, W. H. Melhuish, R. J. Sparks, G. Wallace,C. A. M. Brenninkmeijer, and R. C. McGill (1990), The use of radiocar-bon measurements in atmospheric studies, Radiocarbon, 32(1), 37–58.

Manning, M. R., D. C. Lowe, R. C. Moss, G. E. Bodeker, G. Wallace, andW. Allan (2005), Short-term variations in the oxidizing power of theatmosphere, Nature, 436, 1001–1004.

Marland, G., T. A. Boden, and R. J. Andres (2007), Global, regional, andnational fossil fuel CO2 emissions, in Trends: A Compendium of Dataon Global Change, http://cdiac.ornl.gov/trends/emis/overview_2007.html, Carbon Dioxide Inf. Anal. Cent., Oak Ridge Natl. Lab., U.S.Dep. of Energy, Oak Ridge, Tenn.

Marland, G., T. A. Boden, and R. J. Andres (2008), Global, regional,and national fossil fuel CO2 emissions (1751–2008), in Trends: ACompendium of Data on Global Change, http://cdiac.ornl.gov/trends/emis/overview_2008.html, Carbon Dioxide Inf. Anal. Cent., Oak RidgeNatl. Lab., U.S. Dep. of Energy, Oak Ridge, Tenn.

Marquis, M., and P. Tans (2008), Climate change: Carbon crucible,Science, 320(5875), 460–461.

Masarie, K. A., and P. P. Tans (1995), Extension and integration of atmo-spheric carbon dioxide data into a globally consistent measurementrecord, J. Geophys. Res., 100(D6), 11,593–11,610.

Masarik, J., and J. Beer (1999), Simulation of particle fluxes and cosmo-genic nuclide production in the Earth’s atmosphere, J. Geophys. Res.,104(D10), 12,099–12,112.

Masarik, J., and J. Beer (2009), An updated simulation of particle fluxes andcosmogenic nuclide production in the Earth’s atmosphere, J. Geophys.Res., 114, D11103, doi:10.1029/2008JD010557.

Meijer, H. A. J., H. M. Smid, E. Perez and M. G. Keizer (1996), Isotopiccharacterisation of anthropogenic CO2 emissions using isotopic andradiocarbon analysis, Phys. Chem. Earth, 21(5–6), 483–487.

Naegler, T., and I. Levin (2006), Closing the global radiocarbon budget1945–2005, J. Geophys. Res., 111, D12311, doi:10.1029/2005JD006758.

Naegler, T., and I. Levin (2009), Biosphere-atmosphere gross carbonexchange flux and the d13CO2 and D14CO2 disequilibria constrained bythe biospheric excess radiocarbon inventory, J. Geophys. Res., 114,D17303, doi:10.1029/2008JD011116.

Naegler, T., P. Ciais, K. Rodgers, and I. Levin (2006), Excess radiocarbonconstraints on air-sea gas exchange and the uptake of CO2 by the oceans,Geophys. Res. Lett., 33, L11802, doi:10.1029/2005GL025408.

Nakada, A., T. Miyauchi, K. Akiyama, T. Momose, T. Kozawa, T. Yokota,and H. Ohtomo (2008), Radioactive airborne effluent discharged fromTokai reprocessing plant (1998–2007), technical report, Japan At. EnergyAgency, Tokai-mura, Japan.

Neftel, A., H. Friedli, E. Moor, H. Lötscher, H. Oeschger, U. Siegenthaler,and B. Stauffer (1994), Historical CO2 record from the Siple Station icecore, in Trends: A Compendium of Data on Global Change, http://


13 of 14

cdiac.ornl.gov/trends/co2/siple.html, Carbon Dioxide Inf. Anal. Cent.,Oak Ridge Natl. Lab., U.S. Dep. of Energy, Oak Ridge, Tenn.

Nisbet, E. (2005), Emissions control needs atmospheric verification,Nature, 433(7027), 683.

Nydal, R., and K. Lövseth (1965), Distribution of radiocarbon from nucleartests, Nature, 206(4988), 1029–1031.

Nydal, R., and K. Lövseth (1983), Tracing bomb 14C in the atmosphere1962–1980, J. Geophys. Res., 88(C6), 3621–3642.

Nydal, R., A. Brenkert, and T. Boden (1998), Carbon-14 measurementsin surface water CO2 from the Atlantic, Indian and Pacific Oceans,1965–1994, NDP-057A, Carbon Dioxide Inf. Anal. Cent., Oak RidgeNatl. Lab., U.S. Dep. of Energy, Oak Ridge, Tenn.

O’Brien, K. (1979), Secular variations in the production of cosmogenicisotopes in the earth’s atmosphere, J. Geophys. Res., 84(A2), 423–431.

Oeschger, H., U. Siegenthaler, U. Schotterer, and A. Gugelmann (1975), Abox diffusion model to study carbon dioxide exchange in nature, Tellus,27, 168–192.

Proctor, I. D., J. R. Southon, M. L. Roberts, J. C. Davis, D. W. Heikkinen,T. L. Moore, J. L. Garibaldi, and T. A. Zimmerman (1990), The LLNLion source—Past, present and future, Nucl. Instrum. Meth. Phys. Res. B,52(3–4), 334–337.

Rafter, T. A. (1955), 14C variations in nature and the effect on radiocarbondating, N. Z. J. Sci. Technol., Sect. B, 37(1), 20–38.

Rafter, T. A., and G. J. Fergusson (1957), “Atom Bomb Effect”—Recentincrease of Carbon-14 content of the atmosphere and biosphere, Science,126(3273), 557–558.

Randerson, J. T., I. G. Enting, E. A. G. Schuur, K. Caldeira, and I. Y. Fung(2002), Seasonal and latitudinal variability of troposphere D14CO2: Postbomb contributions from fossil fuels, oceans, the stratosphere, and the ter-restrial biosphere, Global Biogeochem. Cycles, 16(4), 1112, doi:10.1029/2002GB001876.

Randerson, J. T., C. A. Masiello, C. J. Still, T. Rahn, H. Poorter, andC. B. Field (2006), Is carbon within the global terrestrial biospherebecoming more oxidized? Implications for trends in atmospheric O2,Global Change Biol., 12(2), 260–271.

Riley, W. J., D. Y. Hsueh, J. T. Randerson, M. L. Fischer, J. G. Hatch,D. E. Pataki, W. Wang, and M. L. Goulden (2008), Where do fossil fuelcarbon dioxide emissions from California go? An analysis based on radio-carbon observations and an atmospheric transport model, J. Geophys.Res., 113, G04002, doi:10.1029/2007JG000625.

Rozanski, K., I. Levin, J. Stock, and R. E. Guevara Falcon (1995), Atmo-spheric 14CO2 variations in the equatorial region, Radiocarbon, 37,509–516.

Sabine, C. L., et al. (2004), The oceanic sink for anthropogenic CO2,Science, 305(5682), 367–371.

Schneider, M., and Y. Marignac (2008), Spent nuclear fuel reprocessing inFrance, Res. Rep. 4, Int. Panel on Fissile Mater., Princeton, N. J.

Siegenthaler, U. (1983), Uptake of excess CO2 by an outcrop-diffusionmodel of the ocean, J. Geophys. Res., 88(C6), 3599–3608.

Siegenthaler, U., and H. Oeschger (1987), Biospheric CO2 emissions duringthe past 200 years reconstructed by deconvolution of ice core data, Tellus,Ser. B, 39, 140–154.

Stephens, B. B., R. F. Keeling, M. Heimann, K. D. Six, R. Murnana, andK. Caldeira (1998), Testing global ocean carbon cycle models usingmeasurements of atmospheric O2 and CO2 concentration, GlobalBiogeochem. Cycles, 12, 213–230.

Stuiver, M., and H. A. Polach (1977), Discussion: Reporting of 14C data,Radiocarbon, 19(3), 355–363.

Stuiver, M., and P. D. Quay (1981), Atmospheric 14C changes resultingfrom fossil fuel CO2 release and cosmic ray flux variability, Earth Planet.Sci. Lett., 53, 349–362.

Stuiver, M., P. J. Reimer, and T. F. Braziunas (1998), High-precision radio-carbon age calibration for terrestrial and marine samples, Radiocarbon,40(3), 1127–1151.

Suess, H. E. (1955), Radiocarbon concentration in modern wood, Science,122, 415–417.

Sweeney, C., E. Gloor, A. R. Jacobson, R. M. Key, G. McKinley, J. L.Sarmiento, and R. Wanninkhof (2007), Constraining global air-seagas exchange for CO2 with recent bomb 14C measurements, GlobalBiogeochem. Cycles, 21, GB2015, doi:10.1029/2006GB002784.

Tans, P. P., A. F. M. de Jong, and W. G. Mook (1979), Natural atmospheric14C variation and the Suess effect, Nature, 280(5725), 826–828.

Trumbore, S. (2000), Age of soil organic matter and soil respiration: Radio-carbon constraints on belowground C dynamics, Ecol. Appl., 10(2),399–411.

Turnbull, J. C., J. B. Miller, S. J. Lehman, P. P. Tans, R. J. Sparks, andJ. Southon (2006), Comparison of 14CO2, CO, and SF6 as tracersfor recently added CO2 in the atmosphere and implications for bio-logical CO2 exchange, Geophys. Res. Lett., 33, L01817, doi:10.1029/2005GL024213.

Turnbull, J. C., S. J. Lehman, J. B. Miller, R. J. Sparks, J. Southon, andP. P. Tans (2007), A new high precision 14CO2 time series for NorthAmerican continental air, J. Geophys. Res., 112, D11310, doi:10.1029/2006JD008184.

Turnbull, J. C., et al. (2009a), Spatial distribution of D14CO2 acrossEurasia: Measurements from the TROICA-8 expedition, Atmos. Chem.Phys., 9, 175–187.

Turnbull, J., P. Rayner, J. B. Miller, T. Naegler, P. Ciais, and A. Cozic(2009b), On the use of 14CO2 as a tracer for fossil fuel CO2: Quantifyinguncertainties using an atmospheric transport model, J. Geophys. Res.,114, D22302, doi:10.1029/2009JD012308.

Turnbull, J. C., et al. (2011), Assessment of fossil fuel carbon dioxide andother anthropogenic trace gas emissions from airborne measurementsover Sacramento, California in spring 2009, Atmos. Chem. Phys., 11,705–721, doi:10.5194/acp-11-705-2011.

UK Environmental Agency (2008), Radioactivity in food and the envi-ronment, 2007, RIFE 13, Rotherham, U. K.

United Nations Scientific Committee on the Effects of Atomic Radation(2000), Sources and effects of ionizing radiation, technical report,Vienna. [Available at http://www.unscear.org/.]

van der Laan, S., U. Karstens, R. E. M. Neubert, I. T. van der Laan-Luijkx,and H. A. J. Meijer (2010), Observation-based estimates of fossil fuel-derived CO2 emissions in the Netherlands usingD14C, CO and 222Radon,Tellus, Ser. B, 62(5), 389–402, doi:10.1111/j.1600-0889.2010.00493.x.

Wilks, D. S. (1995), Statistical Methods in the Atmospheric Sciences: AnIntroduction, Academic, San Diego, Calif.

H. D. Graven and R. F. Keeling, Scripps Institution of Oceanography,University of California, San Diego, 9500 Gilman Dr., La Jolla, CA92093-0244, USA. ([email protected]; [email protected])T. P. Guilderson, Center for Accelerator Mass Spectrometry, Lawrence

Livermore National Laboratory L-397, 7000 East Ave., Livermore, CA94550, USA. ([email protected])


14 of 14

observations of radiocarbon in co2 at la jolla, california...

Documents